Solvation pressure in ethanol by molecular dynamics simulations

Abstract

The results of all-atom molecular dynamics simulations of ethanol liquid and vapor using a modified version of the Cornell field ͓W. D. Cornell and P. Cieplak, J. Am. Chem. Soc. 117, 5179 ͑1995͔͒ are presented. Excellent agreement with experiment is obtained for density, compressibility, and cohesive energy density. The ethanol liquid is subjected to uniform hydrostatic pressure in the range −1 to 15 kbar at room temperature and the vibrational frequency spectra are calculated. The peak frequencies of seven major vibrational modes are found to be accurate to within 100 cm −1 of their experimental positions and the change of frequency as a function of pressure is consistent with Raman data. The change in bond length is found to be consistent with the solvation pressure model for all bonds except for O-H due to hydrogen bonding.

Solvation pressure in a Lennard-Jones fluid by molecular dynamics simulation

“…While ostensibly a simple system, our results exhibited rich behavior similar to that observed experimentally, using Raman spectroscopy, for a range of solute-solvent systems including those where hydrogen bonding is known to be a dominant factor in bond length behavior. [8][9][10] We found that for x ϳ 1, the observed bond behavior is consistent with the solvation pressure model but with the caveat that neither the effective local CED nor the axial pressure indicate that the pressure experienced by the dimer was that expected for compliance with the model.…”

“…For example, in real systems where strong attractive forces such as hydrogen bonding are known to be present, comparisons between Raman spectra from experiment and simulation indicate bond compression and an increase in bond length as a function of pressure. [8][9][10] We now seek to establish why agreement with the solvation pressure model occurs for a small range of x and why, for large relative attractive potentials, an increase in bond length is observed with increasing applied pressure.…”

“…An increase in dimer bond length with applied pressure is indicative of strong, attractive forces and local structure and is consistent with experimental observations in hydrogenbonded systems. 9 Although the main aim of this study is to test the validity of the solvation pressure model, and hence ubiquitous nature of solvation pressure, we emphasize that all the conclusions drawn have been tested on the LJ system only. Other work worth noting is a study of how the attractive potential influences the structuring of spheres which demonstrated that cohesion between atoms is enormously important to structure and is of greater significance than first realized.…”

“…This leads to changes in the solvation pressure experienced by solvated molecules and good evidence for the solvation pressure effect has been observed using Raman spectroscopy for nanosized objects such as carbon nanotubes, 6,7 starch grains, 8 pure solvents, 9,10 and for solvent mixtures. 11,12 The solvation pressure model asserts that all solutes, including particles immersed in their own liquid, experience a pressure equal to the cohesive energy density ͑CED͒ of the solvent, which is defined as the energy of vaporization per unit volume.…”

“…For example, MD simulations of pure ethanol report that the change in bond length with pressure is well described by the solvation pressure model for all bonds except for O-H, which is lengthened with applied pressure due to masking by hydrogen bonding forces. 9 In addition, a comparison between experimental data and MD simulations demonstrate that masking interactions exist in real chloroform which can be reduced or eliminated in simplified simulations. 10 The different length behaviors observed have left workers cautious as to the general applicability of the model to hydrogen-bonded systems, where the vibrational properties of bonds are sensitive to local structure.…”